Abstract People with latent tuberculosis (TB) infection (LTBI) are at risk of developing active disease. TB chemotherapy often fails to sterilize Mycobacterium tuberculosis (Mtb), leaving individuals at risk of relapse TB. How Mtb survives during paucibacillary persistent TB infections in humans or in animal models is for the most part unknown. Drug treatment and silencing of in vivo essential genes can cure acute and chronic Mtb infections in mice to the extent that CFU can no longer be detected on agar plates. However, as in humans, Mtb in these mice is often not sterilized, and paucibacillary persistence may eventually result in relapse of TB. We have established paucibacillary mouse models by infecting mice with mutants conditionally lacking proteins that are required for in vivo growth and high-titer persistence. We will use conditional knockdown mutants to identify the metabolic pathways that Mtb requires to establish and maintain paucibacillary persistence in mice. Enzymes that are required during growth in vivo, during high titer persistence in chronic mouse infection, and are also essential during paucibacillary persistence represent the most valuable targets. Our work can identify and prioritize them. We will also investigate the importance of key host immune factors for the control of paucibacillary TB and investigate the immune responses associated with TB relapse. This project will lead to a deeper understanding of Mtb?s metabolic adaptations that facilitate paucibacillary persistence in mice, increase understanding of host immune factors that control paucibacillary Mtb and uncover mechanisms of control or disease progression in TB.